Differences

This shows you the differences between two versions of the page.

--- edp [2026/08/16 16:52] – [Step 2:Automatically Get EDP Data] edit
+++ edp [2026/08/27 18:29] (current) – edit
@@ Line 8: / Line 8: @@
 Updated (2023): Abhi Senthilkumar (Python scripts to automatically rename atoms and combine .sim files into xlsx)\\
 Updated (2023): Joshua Lucker (Clarification on Step 6)\\
-Updated (2026): Eric Zhang (Completely Refactored and Automated old scripts) IN PROGRESS\\
+Updated (2026): Eric Zhang (Completely Refactored and Automated old scripts)\\
 Example of analysis: {{ ::e.g._thickness.xlsx |}}
@@ Line 22: / Line 22: @@
 Also if there is a protein download the {{ :protein_analysis.tar |}} files and run them on as many proteins as there are in the system, and combine the data similarly to before
+The initial instructions below are for the legacy scripts. Continue scrolling to see the updated workflow.
@@ Line 284: / Line 286: @@
 Please refer to Excel file eg_thicknesses.xlsx for the example measurement.
-======Instructions Modern Python Workflow (IN PROGRESS) ======
+======Instructions Modern Python Workflow ======
 Many of the steps in the original symmetric workflow have been combined and/or automated. This works for symmetric or asymmetric systems. It automatically computes thickness and interdigiation, so DO NOT do the separate interdigitation analysis by Eric Wang. If something isn't clear, reach out to ezhang32@umd.edu.
 ===== Step 1: Edit Necessary Files =====
-a) Open ''config.py''. Read the contents and fill in the resns list, workspace fields, and analysis fields as directed in the comments. Any patched / conjugated lipids should be entered using a tuple containing the constituent residues. Lipids uniquely identified by their segid instead of resid should also be marked using a single-element tuple.
+a) Open ''config.py''. Read the comments I left in the file; they explain everything in detail. Then fill in config_dict, workspace fields, and analysis fields as directed.
-Example: AIPE and AMPG are standard CHARMM lipids with segid MEMB. If I had a custom lipid composed of ORN and LLL residues, here is how I would populate ''config.py''
+Example: AIPE and AMPG are standard CHARMM lipids with segid MEMB. If I had a patched lipid composed of ORN and LLL residues with segids NAAA1, NAAA2, etc., here is how I would populate ''config.py''
 <code>
-resns = ['AIPE', 'AMPG', ('ORN','LLL'),] # conform to Python syntax by surrounding strings with double or single quotes
+# resnames should be all caps (matches case in toppar files)
+config_dict = {
+    'MEMB':{
+        'AIPE': (58.3686, 58.2105),
+        'AMPG': (61.0204, 59.9090),
+    },
+    'NAAA':{ # for patched / conjugated lipid, set subresidues to have the same comp_sa
+        'ORN': (52.8182, 52.8821),
+        'LLL': (52.8182, 52.8821),
+    },
+    'IONS': {
+        'POT': None,
+        'CLA': None,
+    },
+    'TIP3':{
+        'TIP3': None,
+    },
+}
 </code>
-b) (OPTIONAL) If you have any ions besides pot, cla, or cal (potassium, chloride, sodium), or you have custom/mutated lipids, you'll need to edit ''atmlbl.py'' and ''cmp.py'' inside the src folder. This can get complicated, so study the examples/comments I've already left in the files and mirror the logic for your lipid. Note: the current setup only works for all default CHARMM glycerophospholipids, sphingolipids, and common 4/5/6-chained lipid A.
+b) (OPTIONAL) If you have any ions besides pot, cla, or cal (potassium, chloride, sodium), or you have custom/mutated lipids, you'll need to edit ''atmlbl.py'' and ''cmp.py'' inside the src folder. New atom types + electron weights need to be added to the ATOM_EDP_DICT inside ''4_calc_edp.py''. This can get complicated, so study the examples/comments I've already left in the files and mirror the logic for your lipid. Note: the current setup only works for all default CHARMM glycerophospholipids, sphingolipids, and common 4/5/6-chained lipid A.
 c) Now run ''./0_setup.py''. Confirm that all files are generated successfully before continuing.
-===== Step 2:Automatically Get EDP Data =====
+===== Step 2: Automatically Get EDP Data =====
-a) Run ''1_recenter.scr''. This should take <5 minutes. Confirm that you have recentered dcds in your recenter directory after the job finishes.
+a) If you set recenter_dir = None in config.py, you will see a new recenter script and a directory be created. If not, skip to the next step. Run ''./recenter.scr''. This should take <5 minutes. Confirm that you have recentered the dcds in your recenter directory after the job finishes.\\
-b) Run ''run_edp.scr''. This completely automates steps 2-7 of the edp pipeline. You should see printed output explaining which steps are running. It will take <2 minutes. If everything went smoothly, you're essentially done! You can continue modifying plots by examining the comments in ''plot_edp.py'' or directly tweaking the code in ''src/plot_params.py''. If you set EXP_format = True in ''config.py'', you'll see a new directory called SimToEXP which contains the formatted files required for SimToEXP software. Details about the usage for SimToEXP are covered in Part V of the legacy scripts above, and it's not necessary for most. The results folder contains EDP_results.txt, which holds thickness and interdigitation results plus any plots.
-Note: If you want to rerun steps individually, just run the numbered .scr or .py scripts. Verify that your lipids are assigned to the correct leaflet based on printed output or by checking ''memb_compositon.csv'' inside the sim_data folder.
+b) Run ''1_atm_dens.scr''. The first time you run it, it will take a couple of minutes to set up a virtual environment with the MDAnalysis package installed. Then, it will automate all other steps and run for < 1 minute. You should now see the default plots, thickness data, and interdigitation data in the results folder. If you need NSLD data, all of the formatted .sim and .cmp files can be found inside the SimToEXP folder. Read more about SimToEXP in part VI of the legacy script instructions.
 ===== Additional Information =====
-This section explains the Python workflow in a bit more detail for interested viewers or those encountering bugs.
+This section explains the Python workflow in a bit more detail. I recommend you at least skim over it.
+=== Overall Pipeline ===
+_setup.py roughly validates the structure of all the fields in config.py before generating necessary files. 1_atm_dens.scr create a virtual Python environment, then runs 1_atm_dens.py to get .sim files with atom densities -> 2_calc_edp.py which applies the electron weights and groups atoms according to the .cmp files -> 3_analysis.py which computes thickness and interdigitation data -> 4_plot_edp.py which creates generic plots based on the edp data -> 5_exp_format.py which formats .cmp and .sim files to be usable in SimToEXP.
+=== How to Interpret Last 2 Columns in Thickness Results ======
+^ Metric ^ Rho_Max_or_Neg ^ Rho_Max_or_Mid_Pos ^ What it measures ^
+| **DHH** | //ρ//<sub>max</sub> (Bot Leaflet) | //ρ//<sub>max</sub> (Top Leaflet) | Peak headgroup densities |
+| **DB** | //ρ//<sub>max</sub> (Water) | //ρ//<sub>mid</sub> (Water) | Water phase peak & half-max threshold |
+| **2DC** | //ρ//<sub>max</sub> (Carbon Core) | //ρ//<sub>mid</sub> (Carbon Core) | Core peak & half-max threshold |
+=== Formulas for Self and Hetero Interdigitation ===
+{{:new_interdig_formulas.png}}